Initiating Cardswap Based on Analytic Model Indicating Third Party Card Reissuance

1. A system for initiating a cardswap, the system comprising: one or more computing devices, wherein the one or more computing devices comprises: a memory to store instructions; and processing circuitry, coupled with the memory, operable to execute the instructions, that when executed, cause the processing circuitry to: receive customer transaction data associated with a customer; determine, via a machine learning model, whether a reissue event has occurred, likely occurred, or will occur for the customer based at least in part on the customer transaction data, wherein the reissue event is a reissuance of a third-party card associated with the customer; initiate, based on the determination, the cardswap by identifying one or more cards associated with the customer, wherein the one or more cards includes one or more of the following: (i) a customer credit card, (ii) a customer debit card, and (iii) a customer virtual card, wherein the identified one or more cards do not include third-party cards; present the identified one or more cards to the customer; receive a selection from the customer of a card from the presented one or more cards; identify, based on the customer transaction data, a plurality of third-party websites or web applications where the third-party card is stored to process payments; and automatically perform the cardswap based on the selection of the card, wherein the performance of the cardswap comprises executing a script to: navigate, by a web browser, to each identified third-party website or web application; and replace, by the web browser based on a determination that each of the plurality of third-party websites or web applications includes payment token information, the third-party card on file at each identified third-party website or web application with the selected card by populating an account number of the selected card into a respective form field on each identified third-party website or web application, and wherein the machine learning model is trained using at least transaction activity information associated with customer cards after customer third-party card reissuances, the transaction activity information including at least: use of the customer cards at a frequency after dates corresponding to the customer third-party card reissuances that is greater than a frequency of use of the customer cards before the dates corresponding to the customer third-party card reissuances, and use of the customer cards at new merchants after the dates corresponding to the customer third-party card reissuances.

2. The system of claim 1 , wherein the customer virtual card is a virtual number associated with a customer credit card, a customer debit card, or a customer account, wherein the performance of the cardswap further comprises executing the script to: determine another third-party website or web application does not include payment token information; and refrain from populating the account number into the form field on the another third-party website or web application based on the determination that the another third-party website or web application does not include payment token information.

3. The system of claim 2 , wherein the machine learning model includes a classification model, wherein the script is executed by a browser extension of the web browser.

4. The system of claim 1 , wherein the machine learning model is further trained using information related to: (i) changes in spending patterns, (ii) changes in spending volume (iii) recurring charges on the customer cards, and (iv) placing the customer cards on file with one or more companies by the customers.

5. The system of claim 3 , wherein the classification model is a logistic regression model, a decision tree model, a random forest model, or a Bayes model.

6. The system of claim 3 , wherein the classification model is based on a convolutional neural network (CNN) algorithm, a recurrent neural network (RNN) algorithm, or a hierarchical attention network (HAN) algorithm.

7. The system of claim 1 , wherein the identified one or more cards are presented to the customer via an e-mail, a text message, a call, a notification message, and/or a mobile banking application message.

8. The system of claim 1 , wherein the customer transaction data includes one or more of the following: (i) a purchase, (ii) a financial transaction, (iii) a withdrawal, (iv) a deposit, (v) a recurring charge, and (vi) a one-time charge.

9. The system of claim 8 , wherein the processing circuitry is further caused to determine which one of the identified one or more cards are to be presented for the cardswap.

10. The system of claim 1 , wherein the one or more cards are associated with a first company, and wherein the third-party card is associated with a second company different than the first company.

11. The system of claim 10 , wherein the first company is a financial company and the second company is a competitor of the financial company.

12. The system of claim 1 , wherein the cardswap is performed on or for one or more of the following: (i) an online retailer, (ii) an online media-service provider, (iii) an online subscription, (iv) an online newspaper or periodical, (v) a social media service, (vi) an online gaming service, (vii) a ridesharing service, (viii) a hospitality service, and (ix) an online food ordering service.

13. A non-transitory computer-readable storage medium storing computer-readable program code executable by a processor to: determine, via a machine learning model, whether a reissue event has occurred, likely occurred, or will occur for a customer based at least in part on customer transaction data, wherein the reissue event is a reissuance of a third-party card associated with the customer; identify, based on the determination, one or more cards associated with the customer, wherein the one or more cards includes one or more of the following: (i) a customer credit card, (ii) a customer debit card, and (iii) a customer virtual card; send the customer a message or a notification for swapping the third-party card with the identified one or more cards; identify, based on the customer transaction data, a plurality of third-party websites or web applications where the third-party card is stored to process payments; and automatically perform a cardswap based on a response to the message or the notification from the customer selecting of a card from the identified one or more cards, wherein the performance of the cardswap comprises executing a script to: navigate, by a web browser, to each identified third-party website or web application; and replace, by the web browser based on a determination that each of the plurality of third-party websites or web applications includes payment token information, the third-party card on file at each identified third-party website or web application with the selected card by populating an account number of the selected card into a respective form field on each identified third-party website or web application, and wherein the machine learning model is trained using at least transaction activity information associated with customer cards after customer third-party card reissuances, the transaction activity information including at least: use of the customer cards at a frequency after dates corresponding to the customer third-party card reissuances that is greater than a frequency of use of the customer cards before the dates corresponding to the customer third-party card reissuances, and use of the customer cards at new merchants after the dates corresponding to the customer third-party card reissuances.

14. A method, comprising: receiving, by a processor of a device, customer transaction data associated with a customer; determining, via a machine learning model executing on the processor, whether a reissue event has occurred, likely occurred, or will occur for the customer based at least in part on the customer transaction data, wherein the reissue event is a reissuance of a third-party card associated with the customer; initiating, by the processor based on the determination, a cardswap by identifying one or more cards associated with the customer, wherein the one or more cards includes one or more of the following: (i) a customer credit card, (ii) a customer debit card, and (iii) a customer virtual card; presenting, by the processor, the identified one or more cards to the customer; receiving, by the processor, a selection from the customer of a card from the presented one or more cards; identifying, by the processor based on the customer transaction data, a plurality of third-party websites or web applications where the third-party card is stored to process payments; and automatically performing the cardswap by the processor based on the selection of the card, wherein the performance of the cardswap comprises executing a script to: navigate, by a web browser executing on the browser, to each identified third-party website or web application; and replace, by the web browser based on a determination that each of the plurality of third-party websites or web applications includes payment token information, the third-party card on file at each identified third-party website or web application with the selected card by populating an account number of the selected card into a respective form field on each identified third-party website or web application, and wherein the machine learning model is trained using at least transaction activity information associated with customer cards after customer third-party card reissuances, the transaction activity information including at least: use of the customer cards at a frequency after dates corresponding to the customer third-party card reissuances that is greater than a frequency of use of the customer cards before the dates corresponding to the customer third-party card reissuances, and use of the customer cards at new merchants after the dates corresponding to the customer third-party card reissuances.

15. The method of claim 14 , wherein the customer virtual card is a virtual number associated with a customer credit card, a customer debit card, or a customer account, wherein the performance of the cardswap further comprises executing the script to: determine another third-party website or web application does not include payment token information; and refrain from populating the account number into the form field on the another third-party website or web application based on the determination that the another third-party website or web application does not include payment token information.

16. The method of claim 15 , wherein the machine learning model includes a classification model, wherein the script is executed by a browser extension of the web browser.

17. The method of claim 16 , wherein the machine learning model is further trained using information related to: (i) changes in spending patterns, (ii) changes in spending volume (iii) recurring charges on the customer cards, and (iv) placing the customer cards on file with one or more companies by the customers.

18. The method of claim 17 , wherein the classification model is a logistic regression model, a decision tree model, a random forest model, or a Bayes model.

19. The method of claim 14 , wherein the machine learning model is further trained using information related to: (i) changes in spending patterns, (ii) changes in spending volume (iii) recurring charges on the customer cards, and (iv) placing the customer cards on file with one or more companies by the customers.

20. The method of claim 19 , further comprising: identifying, by the machine learning model, the one or more cards associated with the customer based on: (i) a respective balance of each of the one or more cards, (ii) a respective amount of available credit of each of the one or more cards, (iii) whether each of the one or more cards were previously used for a previous cardswap operation, and (iv) a respective expiration date of each of the one or more cards.